1. Field of the Invention
The present invention relates to an image pickup apparatus that includes an image pickup element having an image pickup pixel and a focus detection pixel.
2. Description of the Related Art
Previously, in an image pickup apparatus that uses an image pickup element such as a CMOS sensor, shading (an offset amount) in a dark situation that is generated in taking an image is reduced by performing a predetermined correction processing. Japanese Patent Laid-Open No. 2003-333434 discloses a horizontal dark shading correction that is performed in a developing processing using one-dimensional correction data that are obtained by performing a projection calculation for an image taken in the dark situation.
In order to improve the performance of the image pickup apparatus, a configuration that does not need an AF sensor for exclusive use by providing a phase difference detection function to the image pickup element so as to achieve a high-speed phase difference AF is known. Japanese Patent Laid-Open No. 2010-20055 discloses an image pickup apparatus that includes an image pickup element having common pixels (image pickup pixels) and focus detection pixels. This image pickup apparatus has an addition readout mode such as an electronic view finder mode or a moving picture shooting mode in which outputs of the plurality of image pickup pixels are added to be read. Japanese Patent Laid-Open No. 2010-20055 discloses an image pickup apparatus that has a mode of performing an addition processing for the image pickup pixels and reading only the focus detection pixel without performing the addition processing (with a non-addition processing) when the focus detection pixel is contained.
However, when pixels such as the image pickup pixels (addition pixels) and the focus detection pixels (non-addition pixels) that are read using different drive patterns are mixed in one frame, a difference of characteristics is generated by the influence of a circuit configuration or the like. For example, when an electric circuit that performs the addition processing exists inside the image pickup element, a switch element such as a transistor is disposed between pixels to be added. On the other hand, since the addition processing is not performed for an output of the focus detection pixel (non-addition), any switch element is not disposed between the focus detection pixels. Therefore, in accordance with the influence of a variation of an individual threshold value of each switch, the shading (the offset amount) is different between at the addition time and at the non-addition time. Furthermore, since the focus detection pixel performs a pupil division, an opening by a wiring layer is different from that of the common pixel. Therefore, a difference is generated in a wiring capacity and the offset amount is different between the common pixel and the focus detection pixel. Accordingly, even when a shading correction (an offset correction) is performed for the output of the common pixel (the image pickup pixel), it is not necessarily appropriate for the output of the focus detection pixel, and therefore an image quality may be deteriorated.